COMPARATIVE MANAGEMENT STRUCTURE

AND INSTITUTIONAL PERFORMANCE

IN RURAL BANKING INSTITUTIONS

Cesar G Saldafia

WORKING PAPER SERIES NO 88 16

This paper isalso beingcirculatedas ACPC Working PaperSeries No 88.08 by the Awicultural Credit Policy Council

October 1988

Ph,l,ppme Inst=tutefor Development Studies

ACKNOWLEDGEMENT

We wish to acknowledge the financial support provided by the

Agricultural Credit Policy Council (ACPC), the Philippine Insti-

tute for Development Studies (PIDS), the Ohio State University

(OSU) and the United 8tares Agency for International" Development

(USAID). This study was made possible with the cooperation of

financial institutions included in this study. We cannot

enumerate their names here for confidentiality.

The Author

TABLE OF CONTENTS

I Introduction 1

II. A Framework for Relating Management Structurewith Institutional Performance..:. ..... ; . • 3

III The Survey 8

IV. Review of Related Literature ........... i0

V. Management Structure and Policies of RFIs .... 16

VI Relating Management Structure and Policies:withOperati g Pe formance of RFIs 31n r ,,,,..,,..,

VII. Concluding Remarks ................ 50

References ...................... 57

LIST OF FIGURES AND TABLES

Figure 1. Management, Structure as, a Factor Influencing theImpact of External Forces on RFI Performance ....... 4

Table i. Breakdown of Respondent Banks ..... ..... 9

2. Performance Indicators by Bank Type Average ...... 17

3. Comparative Organization Structure, AverageNigher/Proportion .... ._. .............. 19

4. Pairwise Comparison of organization structure ..... 20

5. Comparative Degree of Decentralization ....... 22

6. _ducation and Ownership of RFI Management . ....... 24

7. Deposit Generation ar_ Number of Respondent RFIs..... 27

8. Loan Administration and Number of Respondent RFIs . . 28

9. Ownership Background of RBs and PDBs........ 31

i0. Managerial Hierarchy and Scope of RFI Services, Mean. . 32

ii. Staff Size and Overhead Cost, Mean ...........

12. Branch Manager's Discretionary Authority on Loans,Mean 35

15. Ownership, Structure and RFI Performance ........ 57

14. Multiple Recession of Return on Assets Ratio onFinancial and Organizational Variables ..... . . . . 59

15. Manpower Deployment Policies and Service Performanceof RFIs 41

16. Management Marketing Policies and Deposit/LoansPerformance .............. 42

17. Savings Campaign and Deposit Performance forRBs and PDBs 45

18. Profitability Performance of RFIs ......... 48

COMPARATIVE MANAGEMENT STRUCTURE AND INSTITUTIONALPERFORMANCE IN RURAL BANKING INSTITUTIONS*

by

C_sar G. Salda_a**,

I. INTRODUCTION•

Various government policies and programs have focused on the• : f, •

ways to expand and increase the quality of bank'i.ng services .in• . ,, ,

the rural areas., Most of these efforts were specifically

directed at financing programs and rpgulatory policies meant to

promote the effective participation,...of basking institutions in

the rural financial system. Because. of their economic importance

*Paper presented during the ACPC-PIDS-OSU sponsored seminar-workshop on "Financial Intermediation in the Rural Sector:Research Results and Policy Issues" held on 26-27 September 1988at the Cuaderno Hall, Central Bank of the Philippines., This ispart of a larger study on..comparative, bank analysis jointly con-ducted by .the Agricultural Credit Policy Council (ACPC), Philip-pine Institute for Development Studies (PIDS), and Ohio StateUniversity (OSU). T.he project was coordinated:by Dr. Mario B.Lamberte (PIDS) and Dr. V. Bruce J. Tolentino (ACPC).

**Price Waterhouse/Joaquin Cunanan & Co. Professor of

Accounting at the College of Business Administrat.ion, Universityof the Philippines.

The views expressed in this.study are those of theauth anddo not•necessarily reflect those of the Institute, ,

2

in therural areas, policymakers needto regularly monitor their

activities and performance,,

The objectives of this study are: to descr,u= a,,u uv

evaluate the observed management structure, policies and

practices of three major types of banking organizations operating

in the 'rural areas. These private banking institutions - rural

banks, private development banks and branches of commercial banks

- represent .the major part of the rural financial system.

Interest in rural financial institutions (RFIs) is due to the

ongoing reforms in the financial system which point towards

allowing for increased free play of private, incentives in

financial markets. In the past, government banks like the

Philippine National Bank and the Development Bank of the

Philippines assumed, a considerable role in rural finance.

In this study, the organizational, a;id management

characteristics of the three; types of institutions are presented

on a comparative basis, evaluated, and then related to their

overall performance. The results are intended to contribute to a

better understanding of how government policies and market

conditions determine the management set-up of these institutions

and influence their operating and financial performance.

A descriptive model of interaction between market,

organizational structure, and policy and institutional

performance is presented in Section 2. This section also covers

the operational definition of these variables including several

hypotheses which are suggested by the descriptive model.

.

-Section. 3 describes the •survey data.. SectiOn, 4 ,.reviews some

government regulatory and policy ini,tiativesWhi;C, •impinge om

management structure and policy choices by RFIs. Related

findings by prior studies are also reviewed. _ec_lon 5 cover8

the observed management structure and policies and analyzes the

data for the three classes of RFIs on an individual and.. ,.

comparative basis. Section 6 extends the analysis to relate1?

management structure and policies with overall financial anti

operating performance of the RFIs. The last section presents

some conclusions of the study andsuggestsareas for extension

and analysis_

II. A FRAMEWORKFOR RELATING MANAGEMENTSTRUCTURE WITHINSTITUTIONAL PERFORMANCE

This study of the management aspects of,RFIs _adopts th_

familiar contingenc y,;model used in business policy and control.

The model simply , states that an organization and management

policies represent _-he firm's•way of innovating to .survive and

grow in the face of,,p_eva,i,ling regu)ation/leg_l structure in the• , .... _ '_ ,.,•

marketplace, in short, management policies artd pra_tices are the

RFI's means of competing. These are methods chosen by management

"to exploit opportunities available in the market while avoiding

constraints of regulations or its own weaknesses (e.g., resource

limitations). Self-interest behavior is at the core of thisI : I !

framework. '_The RFI will seek to maX_'mize its returns (commensu-

1/For example, see Porter (1980) and Soriano (1976).

4

rate with risk) when selecting from •among alternative policies

available to the RFI management;

Management structure consists of organization structure,

staffing, and policies guiding the operations of a bank. The

impact of.regulationsand market changes on bank performance can

be better understood in the context of this management structure

of RFIs. In this study, these management aspects are described

and related to the _erformance of individual RFIs.

The .interaction of management structure with market

conditions and regulations is reflected in a schematic diagram

shown below:

MARKETFACTORS'o InterestRatesoF,ndsBources .....oCmpetitiono EconomicUnits_ RFI'sNANA_MENT INSTITUTIONALPERFORMANCE

intheArea L STRUCTURE OFTHERFI-- oOrganization oReturnonAssets)tructure oReturn.onEquity

o _copeofAuthority oI)epositGenerationoManagementCapacity oLoanstoRural

REGULATORYSTRUCTURI.---)oOwnership..and Borr_rsoMandateofRFIs .ManagmntControl oProd,ctivityofStaffo CreditProgram•oCreditPolicyoSupervisionof ..RFI_

FIGURE 1. MANAGEMENT STRUCTURE AS A FACTORINFLUENCING THE IMPACT OF EXTERNALFORCES ON RFI PERFORMANCE

The organizational and management_policy_structure by RFIs

should be Seen as the main adaptive and adjustment mechanism

available to these institutions. •From a s positive theory

standpoint, the observed management structure of, RFIs •must

reflect the best.mechanisms.available to .RFls,,for coping with •

market and regulatory conditions. It is implied that RFIs will

design their own management structure in accordance with theirc -.

self interest, rather than based on the "equity" objective of the.2/

government. As reported in Tolentino (1987, p. 28), the

government remains concerned with "how to ensure that rural farm

subsidies, when channeled through RFIs, are not captured by the

banker."

A number of implications emerge from the suggested

descriptive framework. One, an RFI will design its organization

and make its staffing •decisions to suit available market• i ..

opportunities. For any given set of market opportunities, it

will seek cost-minimizing choices of organization structure and

staff. Two, the regulatory structure will be primarily perceived

as a constraint to the achievement of cost-minimizing choices.

If there are differences in regulations or in mandates (e.g.,: L . ', ' "

size of capital across the RFI types), the effects should be seen

first in the difference in management structure; overall,_ _..,

performance will be evaluated later. Three, any planned regula-

2/Many RFI managements can claim that profitable results and

growth in the rural market is not possible unless they servicethe agricultural lending needs of the local area. This is likelyto be true in many cases, i.e., the efficiency and equityobjectives need not be inconsistent.

6

tory change, (e.g., directed at "equity" Considerations), should

be evaluated in the context of the ability of the RFI$ to

eventually modify its management process in favor of "efficiency"

or Profits._For any given state of management structure, any such

change in regulatory regime will have an adverse short-run effect

on the efficiency of the RFI.

#

In this study, the current market and regulatory regimes are

taken as given and a descriptive analysis of the management

structure of the three types of RFIs is conducted. Operational

indicators of the key elements of management structure are

identified and utilized.

Certain hypotheses can be generated from the relationship

of management structure and performance, using the preceding

framework. First, an RFI's management can be expected to

optimize, within its legal prerogatives, the use of its

resources. In banking, these are primarily the personnel, the

physical branch network .and the flow of funds. Second,

management policies can be expected to be SO designed as to

minimize the impact of regulatory restrictions that adversely

affect the RFI's funds generation and application. Third, for

RFIs whose owners dominate management, policies are the means forI

resolving potential conflicts among the owner-management group,

the government and the other minority holders.

7

Some preliminary questions on the relationship between

performance and management are presented as follows:

,I_,.If management wants to optimize resource utilization,

what policies will itselect,.for:

a) service>offerings?

b) deployment of bank staff along service lines?

2. How does the RFI choose management policies to minimize

regulatory restrictions and exploit available

" opportunities, specifically those related to:

a) retention of a certain percentage of deposits

for lending in the local area?

b) deposit mobilization?

3L Is there a potential relationship between certain

management policies and the condition whereby

stockholders themselves manage the RFI? Specific

questions which can be addressed are those related to:

a) ,returns to stockholders of the RFI;

b) lending to directors andofficers and its impact

on RFI performance.

The study involves an analysis of the operating, policies of

the three RFI types along certain measurable,, and.,,qualitative

c:,_iteria, The more quantitative indicators, of,_ management

operating policies and practices are covered, in, order to

aggregate and highlight the comparativeaspects of the issue

across the RFI types.

8

III, THE SURVEY

The data used in this study was taken from a primary survey

Of 66 rural banks and branches of development and commercial

banks. A questionnaire and an interview schedule were designed

and pre-tested prior to the actual survey. A breakdown of the

responding banks is shown in Table 1.

The 66 respondent banks are broken down into 23 rural banks,

16 unit and branch development banks, and 27 branches of

commercial banks. These banks are spread out over 10 cities and

25 municipalities located in eight provinces from seven regions

of the country.

The survey instruments used were of three types:

(a) Documents submitted by the bank such as financial

statements, organization chart, personnel data, and

special schedules of selected financial information,

e.g., Directors, Officers, Stockholders and Related

Interests (DOSRI) loans;

(b) A 14-page pre-tested questionnaire in two versions:

one for unit banks and another for branch banks.

Research assistants interviewed respondents to clarify

responses, and to follow-up questions with ,misslng

responses.

(c) A two-page interview schedule for the more difficult

and/or sensitive questions.

Table 1

BREAKDOWN OF RESPONDENT BANKS

Location Rural Development CommercialBanks Banks Banks

Region I

Pangasinan 3 2 4

Region.III

Nueva Ecija 0 4 3

Region IV

Batangas 3 3 3Laguna 0 3 4

Region V

Camarinee Sur, 3 1 2

Region VI

Iloilo 4 1 3

Region VII

Negros Oriental 3 1 3

RegionX

Misamis Oriental 7 1 ,. 5

Total 23 16 27

10

The instruments covered the general ,areas of management

structure, operating policies, and bank operating procedures and

systems, including reporting and-evaluation systems. Balance

sheet and income statements for the period 1981-1986 were also

requested including more detailed schedules for selected

financial variables, e.g., DOSRI, number of deposit ,and loan

accounts.

The questionnaire and interview schedule were generally

well-covered by respondents but corresponding financial data

submitted were incomplete. In particular, it was difficult to

obtain financial statements from commercial bank branches. For

most responding banks, problems were encountered in terms of

differences in reporting formats, sometimes leading to gaps in

line item information.

IV. REVIEW OF RELATED LITERATURE

RFIs operate within the regulatory framework set out by the

1980 financial reforms. Lamberte (1987) reviews :._the

implications of the reforms on the operations and organization of

the RFIs. From an operational viewpoint, the current regulatory

environment reduced the differences between these fina.naial

institutions. However, important differences remain inkey areas

of minimum capitalization, limitation on ownership of

enterprises, branching and reserve requirements. Lamberte ciited

rural banks as facing the most restrictions, particularly on

investments and branching. However, rural banks enjoy lower

reserve requirements compared to commercial and development

11

banks. ;t s@ems clear theft .these regulations were set out with

the entire f.inaDcial system in mind._;.However, see_-..onIyfrom the

subset, of the rural0financ,)al sectory,pthe framework.still allows

for the operation of_,three,::institutionswith vastly dif,ferent

equity ;bases ran.gimg;from a minimum of _500,O90._,f_Fr,ural.:,.banks

and _500 million for unibanks,.. A r_a_ch ques_jo_ugges,ted ,is

whether the diff.erent,ial reserve requirement can iRdeed,,:.offset

this difference in resource base, as pointed outby Lamberte.

The_reduction.of _functtonal.differences a_g ba_k types may

,have been less. effe¢¢ive due to. the_remaining)di_ference= in the

resource base .of banks. Wh-ile development.,-banks can. ,l.egal_l_y

offer services and open branches nationwide,.simiclar co,commercial

banks, the fact that these banks are.,6maller prevent them, from

availing.,;ol,this legal,opportunity. The sa_e..,ca_,be s&jd, .abQut

the.opportunity for rural banks to open.branc_h_,.S,_n a region_w),de

basis. ; The r.e_ponse,,of RFIs, .,._o the)'_-_eregulated market

environment can also be seen in the-same context,. ,Such ba_ks may

sti].l be. unable to.,_njby ,benefits from any economies ,of. scope

.(Lamberte 1987.,., p. 7) potentially: avai,3_a_le through,

deregulation.

Meyer (1987_) ,conjectures that rural,banks may be,_unable

to take up, the task of credit evaluation ._nd lending, in the

absence ,of_ thadit.ional Central. Bank_nitia,L_ed programs,, which

speci._y ,theiP-.ow_ <rules. and regulations,-. Thisca_,be taken as_-a

comme_t,;to, the potentia],, lack of ;.management. capa_,i,ty ,_nd

inadequate organ_zat_qn _f rural bankS... Centra_,.Bank',s a_empt

12

tO 'wean rural lenders frromrediScounting, and speci_1 credit

programs was one. of the .agenda .@or the agricultural s_ctor in the

financial reforms of 1980-1985., Tolenti_o (1987) reviews _he

measures which Central Bank implemented to,' ....correct the

agricultural, portfolio .of banks,. He reports:that fewrUral banR_

parti¢,ipate@_i'n these programs, indicating that the. rural., bunks.'

arrearage@ _ndition still per@ists, at least up to. 1986.

Previous studies on the behavior and performance of RFIs

conclude _ thatt_ese institutions are driven by. profit_@eeking

motives in .responding ,to:market and regulatory constrai_ts._ " In

his review of the causes of the decline in formal _@riCUlCbta1

credit, _'.'.Tolentino,"(1987, p.. 5) concludes, that "'len_ers_. as

'_rofit-maximizing businessmen, , Seek to lend to those seotO,rs

Where their combinedcost of funds and supervision are re,]atively

lower, ;_under given _ _rates.¢harged on loans.". Whenever _legal

restrictions on. lending" interest rates preWented" RFIs .-from

earnihg. reasonable Profits, RFiscan simply stop L makih_ 1OanS-',t'O

agriculture. Even under .thecurrent regime .OF. deregulation,

Tolentino reports .that. risk-and default condi'tions..in " Che

agricultural loans market remain unattractive for lehder@. _

The. RFIs ,_ responses to regulation are 'consiste_it with

profit-seeking' behavior. :A._deposit retention scheme -r@quiF_s

that at 1,east 75 percent of the total deposits generated, net_c of

,reserves, should_become part of the loan portfolio of the RF.I _in

:the sa_e _ area."Lamberte (1987)found that on the bas$s:,',of

regiohally-aggregated portfolios, commercial bank (KB) branches

did not,meet this legal requirement. Here was a case where

13

manage/_nt, pollcie_ apparently, violated exp1_icit..-regu1&tory

restric_i_,@,, On_.the ....other hand , the effects of: regulato.r _,

incentives._@p.pear.more d_fficult to assess. Di.fferential.reserv_

requirements i.n.favor of rural-banksare intende_,.!to,gffset the

cost ,advantages _of l_Fger banks. • Lamberte was_unable to trace

this advantage to any superior profitability of rural banks (RBs)

as compar@d to KB branches.

Some regulatory _sstrictions seem to be less _elevant at the

RFI level.,, . The,. minimum_net worth to risk ,aSsetS. ratio is 8

Bercent @or universal bank@.and 10 percen_ .._iforother banks.

Lamberte (1987, pp. 5-6) noted that the regulation is actually

_an: in_itation.fo.r banks,_to_expand their capital." ...HoweVer, it

could also be a deterrent tO asset expansion and loans by banks.

Note that, ,the ratio i sonly rel.¢vant at the ,unit bank 1eve],,

i,e., at the RBs, development bank (POB) and _KB head, offices

rather than the branch bank. ,For example, Lamberte was unable to

ascertain the "equity" of a KB branch. The,implication is that,

while the regulation affects RBs, it can be effectively ignored

by the management Of KB and PDB branches.

Lamberte. reports the qomparative!.performance.of the three

RFI types on.the basis of reg.i.onal.ly-aggrega_ed_.financial data.

Some of the.. results have a d_rect relationship with the

differential management, policies that;each.RFl..type.appears to

have adopted,, Fi, rst, KB branches' deposit mobilization ,is more

It i@,_ conce._n of head office and the branc_ i.gnores,..itexcept on instruction from the top. " '_

14

extensive than PDBs and.RBs. This finding suggests the.need_.:.:'for

a cl'_oser"look into 'the organization and policies adopte8 by .e&_h

RFI type •regarding deposit mobilization. The._'s_ituation .is

reversed in the..case of loand;: The loans to deposit ratiOS are

much. lower for RB branches than for PDBs and RBs in E_mber_e's

study.,

Various conjectures related to management aspects are

offered relative to the dep6_it.and loans-Pelationships for RFIs.

Lamberte Claim8 '_that KB branch management ma'_ have greater

discretionary authority on ra4sing deposits "but .very T__i"ted

authority for Ori"ginating loans. • Meyer (1987, p_"6) suggests

that the RB may be unable "to. respond to thenew unregulated

environment where deposit, m®bilization is supposed, to 'replace

Central Bank funds." In thefaceof.the apparent vioI_tion'_by

some KB branches of"the deposltretention scheme,. Lamberte noted

that this regulation is'nei_ther strictly followed by R_Is."nor

enforced by the banking authorities.

On this aspect of institutional performance, Lamberte (1987)

found, that for the 1983-85 period, the regionally aggregated

loans of development and.rural banks exceeded 'deposits. •: This

finding highlights the importance of the Central Bank

rediscounting and credit programs to these banks. Management

have less incentive, under that regulatoryregime, .to "devejop

the,capaclty for deposit generation, In contrast, Lamberte noted

that in regions outside Metro Manila, branches of commercial

banks are specializing in deposit generation to support, lending

at their headoffices. These findings suggest certain basic

15

dlfferences _n organlzat_on and management which ire explored in

this study The aggregated data in Lamberte's_study also needs

to be verified and retired to the organization _d management at

the bank level.

Comparisons on the bas_s of profitability turned out to be

more complicated using standard ratios in the Lamberte study

Due to the disparity in deposits and loans KB branches appear

unprofitable . W_thout reasonable transfer price estimates for

fund transfers from branch to head offices, reported revenues of

KB and PDB branches are understated Even the measure used in

Lamberte's study -- net operating income as a ratio of operating

income -- is clearly deficient in this respect Given this

system of accounting, cumulative net branch losses will result

in negative equity for most KB branches This situation

makes return on equity, a traditional efficiency measure,

meaningless when applied to unad3usted branch financial

statement Lamberte suggests that future studies should innovate

on measurement techniques to arrive at the true p_cture of RFI

performance

In summary, the current evidence suggests a menu of adaptive

policy mechanisms that RFI management follows, guided by its

proflt-_eeklng motive Faced with a market condition of income

regulation, the RFI seeks low cost _nternal arrangements If the

regime is one of deregulation, RFIs will seek highest portfolio

Equity eauals assets less l_ab_l]tles of the KB or PDBbranch

16

returns while keeping low-cost sources of funds Agricultural

loans w111 decrease if loan interest rates become too regulated

or if urbanized loans yield more returns under deregulation The

alternat19es open to management in dealing with Lt__£JL

conditions is more d_verse Incentives will e_ther be taken or

_gnored Restrlct_ve regulations may either be followed or

violated, the latter case even with or without penalties

Restrictions may have full impact on the management of unit banks

in the rural areas but have no consequence on competing branches

of regional/national banks _n the same areas RFI management may

be required to simultaneously ad3ust to an incentive (e g , lower

reserve requirement) and a restriction (e g , net worth to risk

assets) Performance results come out of the adaptive responses

of management by way of functional strategies (deposit policy,

loans policy, etc ) However, extra care must be exercised when

using summary lnd_cators like profits due to the lnteract_on of

the measures with unique strategies pursued by RFIs

V MANAGEMENTSTRUCTURE AND POLICIES OF RFIs

1 Mana_emenl_S!_ructure

The basic lndlcator_ of management structure used in the

study are (a) organization structure (b) degree of

decentralization (c) staffing and management background and

(d) ownership Organization structure _s charadterized further

in terms of (1) the number of levels from the bank's chief

operating officer to the bank_staff in face-tO-face contact with

bank client and (11) the number and types of staff poslt_ons in

17

the bank organization The degree of decentralization refers to

the location of approval authority for,deposlt terms, loans to

cllent_ and Investment of bank funds Staffing and management

background covers the mix of managerial and rank and file staff

in the bank and the_educatlonal background of key officers

Finally, _he distinction, _f any, between management and

ownership are explored

These management structure varlables w111 then be related to

selected indicators of bank performance The usual aspects are

size, deposits loans and profitability To provide an initial

reference point, a summary of key performance Indicators are

shown in Tab]e 2

Table 2

PERFORMANCEINDICATORS BY BANK TYPE AVERAGE(In thousand pesos)

..... ===================================================

Developmen_ CommercialRural Bank Bank Bank

1 Bank Size

a) Total Assets PlO 440 J21 823 P70,480b) Premises 97 260 1,372

2 Deposit Generation

a) Amount 3,722 14,406 67,148b) Number of Accounts 4,434 3,772 10,131

m/3 Loans

a) Amount 7,137 10,838 6,165b) Number of Accounts 980 237 63

Excludes outstanding transfers to Head Office for branchesof ¢ommerGlal banks

18

The size, and complexity of a bank's organization structure

depends on bank size_, range of- services, , personnel

expertise and prevail_ing regulatory structure... ' Assuming a

certain degree of mobility of the labor force, personal expertise

.can be contrdlled by a bank through proper hiring and personnel

development policies.. The scope of services offered by all three

types of banks are similar. Only commercial bankbranches offered

other significant bank services such as telegraphic transfers,

drafts/managers' checks, foreign exchange services, and L/Cs.

Nevertheless, the great disparity in asset size and in relative

importance of each banking function makes for organization

structure differences. An extensive branch structure for_

commercial banks can also lead to differences in managerial

authority compared to unit banks like RBs and PDBs with less

branches infar-flung places.

Unit banks..-can be expected.to maintain more organizational

levels compared to branch banks. Unlike unit banks, branches of

Commercial banks can maintain common support services at the

Central o?fice or regional level. Well-developed operating

guidelines may also allow KB branches to operate with

comparatively less need for supervision. .Instead of the value! of

financial assets held by a bank, the number of accounts may bel a

better indicator of an RFI's workload and consequent need fori a

more extended supervisory hierarchy. Table 3 shows a compari'son

of organization structure indicators for. the three .types of RFIs.

Reference can be made to standard-bank management ._ext-books like Johnson and Johnson (1985).

19

Table 3

COMPARATIVE ORGANIZATIOR _ STRUCTURE 'AVERAGE NUMBER/PROPORTION

( Standard Dev i af,'i on)

B A N K SA

Rural Development KB Branch • $_gnificance

1. Number of Levels 4 3.5 3.6 ,N.S.: (0.94) (0.90) (0•52)

2. Plantilla Positions

a) Number 9.15 8.46 8.31 .03(3.36) (3.33) (2.89)

b) Per cent of ' 'Total Personnel 84_ 76% 57_ 0.002

II '(10_) (27_) (25X)3. Hierarchical Mix

a) Proportion of 28_ 27_ 20_ 0.12Managerial/ (8_) (9_) (i3_)SupervisoryPositions toTotal Staff

b) Average Numberof Customer

Accounts Per 3,995 1,985 2,4-50 N.S.Manager/Supervisor (2,336) (1,614-) ' (1',2_1)

Kruekall-Wallie (K-W) One-Way ANOVA (correc_d'for"_es), ,f., _..... _Chi-Square significance: N.S. - Not Slgnl Icant at 0 2

S(Level) - Signific_bt (level)

Among r a sample of 39 RFIs RBs turned i_Out to have the

highest average number of organization levels followed by KB

branches and PDBs. Given the small sample size '_ and unknown

distributional characteristics o? the data, a non-paPametric test

(Kruskall Wall is One-Way ANOVA) was used to compare ,Lthe

distribution of levels across three bank types. The test,". " " • c: *

20

essentially based on comparative ranking, shows no significant

difference in numbe_ of levels, RBs and:PDBs maintain a higher

proportion of distinct positions in their organization compared

to KB branches, This is likely to be explained by the diversity

in primary and support functions in unit banks compared to

reliance... by KB. branches for central support .staff at their

corporate groups. Another possible explanation .may be the

difference in the number of...deposit and loan accounts handled by

unit banks.compared to KB branches.

As in the case when a statist_cal difference, is found, a

pair-wise statistical comparison can be made using another non-

parametric test, the Kolmogorov-Smirnov (K-S) Two-Sample .Test.

The conclusion.regarding the organizational structure, difference

between unitbanks and KB branches is borne out inTable

Table 4

PAIRWIS.E OOMPARIBON OF ORGANIZATION STRUCTURE

I. P'lantilla. PositioD as_/.

Percent of Total Personnel .0023

RB Vs, PDB (N.S.)RB Vs. KB (0,001)PDB Vs. KB (0,08)

2. Hierarchical Mix

Pro.por;tion of Managerial/ ... 12Supervisory Staff to TotalStaff

RB Vs. PDB (N.S.)RB Vs. KB (0.13)PDB Vs. KB (0.14)

L

K-W One-Way ANOVA: All Three Bank Types

K-S Two-Sample Test

21

Table 3 also shows a difference in the hierarchical mix of

these banks' personnel Again, RBs and PDBs have more managerial

and supervisory personnel in contrast to the larger rank and filei/

staff of KB branches This finding indicates lesser control

over . RBs as compared to KB branches There are several

operational explanations in this regard The expertise of the

managerial/supervisory personnel _n the RBs may not be mu_flc_ent

in supervising units with d_verse functions On the other hand,

KBs may s_mply be operating a staff involved in fewer banking

functions compared to unit banks The palrwise comparison _ for

managerial mix _n Table 4 also shows s_gn_ficant statistical

d_fferences along the above direction.

The aspect of decentralization is evaluated 1n the context

of a multi-branch and a unit bank environment O_e way is to

consider the degree of latitude exercised by the manager (of a

unit bank) or the branch manager (of a PDB or KB branch) in loans

approved and decisions on deposit terms Under a _central]zed

authority structure, the manager refers more decisions to a

Committee or President (for unit banks) or to an Area Head br

Central Office (for PDB or KB branches) The results on a

comparative basis are summarized in Table 5l

KB branch managers are allowed much higher loan approval

llmlts than the PDB managers although the latter's llmlts are

Defined as the number of organizational units handled by amanager See Koontz and 0 Donnel (1984)

22

,Table 5.

COMPARATIVE DEGREE OF DECENTRALIZATIONMean (Standard Deviation)

B A N K S

RB DB KB Significance........... w ....

1. Maximum Loan ApprovalLimit of a Manager

Amount (in thousand pesos) P8,667 _43,750 P651,667 0.0001(6,673) (41,908) (837,972)

2. Level of OrganizationWhen Decision on DepositTerms are Made

a) Board 50.0% 18.8_ 4.4_b) President/Top

Management at H.O. - 68.8 69.6c) Manager 18.8 12.5 -

d) Board and Manager 31.2 - -e) Board and H.O. - - 13.0f) "H.O. and Manager - - 13.0

100.0% 100.0_ 100.0_

_, _=_ ......... _

K-W One-Way ANOVA

Z/not far. behind. Disregarding. other factor,s, higher loan

i

approval limits correspond to a "decentralized" set-up., Factors

such as better knowledge of local conditions, availability Of

Z/Interpreted here in a distributional sense. KB (..branches)

has a distribution which is skewed to the right. Most loan l.,imi_sfor KB branch managers are in the range of PDB's except for a fewexceptions with very high limits. One KB reported branches withloan approval limits of _2 million, resulting in the large meanand standard deviation values. Various industry sourcesquestioned the adherence of KBs to the actual excercise of suchlarge approval limits in practice.

23

centrally _ _idm_istere_ gu+delines and faster or more personal

service are some reasons for allowing PDB and KB branch •managers

more discretion on loans, These reasons are normally absent for

accessible branches of PDBs, While PDB and RBRBs and more _ . ._ ._.

managers must pass on loan approval, to a Credit Committee or the

Board, these decisionmaking units .can be readily convened on

short notice,

Furthermore, the degree of discretion allowed to lower

level officer.s for loans may be related to the importance 'given

by top management to the' lending function. Generally, pol-icies

and decisions for the more Critical functions are;e_.pected to be

centralized and if decentralized, are centrally monitored. From

a performance standpoint, it would be intere@ttng to')find out

whether actual average 1cans are within%he, approval limits set

by RFIs as a matter of PolicY.

Deposit policies and decisions are considered top management

prerogative foral.,liRFI._:,.types(Table 5). The cbmpe_i,tiveness of

the RFIs are especially sensitive to deposit terms. Decisions

,rela_ea _o aepos_s are a_so less frequent and closely related to

how the RFI positions itself relative to the market and the

regulatory environment,

As can be expected, the background of RFI managers are

mostly related to'-busin_s or law._ This impl,ies prior',,trai_ing

and familiaPity by RFI managers with, business management .and

legal aspects. A summary of this fi_ndin.gis @i_Swn in Table 6.

24

Results also i,ndica_e that most ,presidents or. man_gere ..Q_, RBs

,ar_.stockholders of their respective banks.

Table 6

•EDUCATION AND OWNERSHIP OF RFI MANAGEMENT

RB. DB KB

-1. .Educational Background ofRFI Managers

a. Business or Law 61.1_ 81.8_ 92.3_

b. Other Fields 38.9 18.2 7.7

2. Ownership and Management:Number of banks or.brancheswhich are

a) Owner-Managed 18 ;.3 O.

b), Not Own.er-Ma_aged -l. IO 24 ,,,

19 ._.3 .. 24

2. Ma_Bqemen_ Po}j:_ies and ODer_tina ,Dec,isi_ons

RFi" management pol-i_ies can b.e systematically described and

compared in terms of the two key functional areas in banking,

namely, deposit generation and loans administration. In

addition, the banking business is concerned with adequacy of

capital and of returns. Bothaspects are joint consequences of

deposit and •loans•policies adopted by management relative to _the

goals, assigned toit by the owners. All these ae,l_ects , are -now

presented in comparative form.,

2S

OeDo_i_ Geh_eration

Table 7 shows relevant aata. ana statistical results for

deposit generation. References to item numbers in Table 3 are

noted in parenthesis.

Except for five RBs, all surveyed RFIs reported that there

are existing written policies on deposits, .specificalt, y, terms

andinterest,-ra£e'scheddles (Item I). Among KB and PDB branches,

such-' de_osit_policies are set by_head office/_oD management. ' A

number of RBS indicated that such polici@s are decided by the

manag@r although majority of RB's point to their Board of_Trustees

(Directors) as the deposits policymaker. Except for six RBsi the

RFIs' deposit performance is 'regularly reviewed (Item 2). The

deposit _view function is one of the main r@spbnsib_lities of

the RFI manager (and jointly with the Area Head in the'case of:KB

branches). Most KB branches carry out this evaluation function

monthly whiYe some branches do it daily (Item 3). _8 _nd PDB

branches :meanwhile, conduct their_review month]'y . _uarterly or

annually;

The nature of deposit campaigns depends on the RFI type.

KBs anchor their campaigns on a motivated internal bank staff by

giving them incentives to contact more people and enlist them as

depositors. On the other hand, KB and PDB branches' campaigns are

comparatively more customer-oriented, usually with raffle prizes

to depositors and TIPID movement schemes. Certain differences

also arise'0n who initiates and how long the deposi_ qampaigns

will be across RFI types. Host RB and PDB deposit campaigns are

26

initiated, by the branch whereas for KB branches, the mandate

comes from the head office (Item 4). Most RB savings campaigns

last "only from one tosix months compared' to six months to

one year for PDB and KB branches (Item 5).

LQan_dm in_ st rat ion

Table 8 summarizes the relevant data and statistical

infer.ences in which the conclusionsin the following p_ragraphs

are based. Reference to items in the table are noted,_ in

parenthesis, Many surveyed RFIs decide to lend based, on written

loan policies. (Item I) set by the Head Office. or Board of

Directors. ' However, a significant number of:rural banks operate

without such written policies, leaving the lending decision to

operating management. The preparation of a loans budget is a

r_gular activity of loans administration for all RFl.types..(Item

2).._ There is a clear diffenence as to the organizational

locatioo, of this activity. PDB and KB branches.cen._ralize ;this

planning, role at the branch.manager or head/are_ office level.

In contrast, RBs delegate the responsibility to lower level staff

including the loan appraiser, credit and collection staff and the

cashier.

•" Except. for RBs where _the modal response to the:_,question:"Who sets loan policies?" is "Central .Bank" (nine out of 22respondents). ..

27

Table 7

:;DEPO_:I,T GENERATION, AND NUH_R OF RESPONDENT RFIs

B A N K Si

RBs PDBs KB Branches Significance

1. __ies .035

a) With Formal/WrittenPolicies 16 16 25

b) No Written. Policies 5 0 ,,_.,

2, Monitorin_ of Deposits ._03

a) Regular Review 15 16 23

b) No Regular Analysis 6 0 0 _,

3, Incentives forDPJ:_osits ,,Generation .09

a) Presence of StaffIncentives 6 8 '_l

b) No Staff Incentives 15 7 10

4. Decisions on Savina_Ca_,paign

a,). Board or Head Office 4 5 1_:

b) Manager 13 5 1_,

c) Other Officers 4 1 . 4,_

5. DuraJ_i_n of SavinasCamDaian

a) One Month z 2 1

b) Three to Six Months 4 2 7_Z "-

c) One Year 1 3 3

d) Other 3 2 1

Chi Square Test. N.S, at 0.2

28

'Table 8

LOAN ADMINISTRATION ANDNUMBER OFRESPONDENT RFIs

================_ ________ .... _ ..... _

S A N K S

ITEM RBs _ PDBS KB Branches Significance

1. Loan Policies _ 0,1

a) With Formal�WrittenPolicies 14 14 21

b) No Written Policies 7 2 2

2, Loan Budqet °N.8_

a) Regularly Prepared 17 13 -21':

b) Not Prepared 5 3 2

3, Ma_ior CompetitorfOr. Loan_

a) Rural Banks 8 5 0

b) PDBs 2 5 4

c) KB branches 4 11 2i

d) Others 2 2 2

4, Loan Collections _._,

a) Incentive to Staff 5 6 8

b) No Incentives 17 9 13

5. Loan Restructuri n_ ' O.'t_

a) Written Guidelines 16 14 18

b) No Formal Guidelines 6 1 1

Chi Square. N.S. at 0,2.

29

Wlth regard ¢o _tltlon in t4_ loans market, many PDB and

most KB branches perceive other KB _ranche_ as their main

competl_ors In !_elr area o_ _perat_ens (Item 3_ Llk@w_se, most

RB an_ man_ PDB branf_he$ perceive _l_er RBs_a_d PDBs as their

co,k_tltOr$ This f_ndlng demonstrates segmemtatlon in the loans

market of RFZs refle_$t_g a dqfference in loaB_market targets or

cllen_ele across RFI types Personal vistts/centacts by.#he bank

staff is regarded by the RFIs as the most effective way of

advertising loan services aside from posters, souvern_rs and

giveaways ,

The _mportance of Joan collection is emphasized by all RFIs

thus, _nce_%jves _n,the form of merlt _ncreas_es and bonuses _re

gIyen for effective_o11@Ctlon efforCs by the bank staff (Item

4) The _raRc_l manager reviews the lean portfolio of all R_Is

regul&rly_ Every RFZ respondent (except one PDB branch)

malntaln_d a syscem fer_monltorlng past due accounts by means of

status re_gorts on ]oam_;outstand_ng Problems on loans are dealt

w_tb based on existing guidelines for loan restrUCtUring (Item

5) However, a s_gn_flcan¢ number of rural barks do not have

such guidelines To help m_n]mlze these contingencies, many RFIs

provide lncent_ves to borrowers for early or prompt loan

repayment through interest rebate, Increased assurance of new

loans and interest d_scount on new loans.

O_ne_shio Background

Management policies of RFIs are related ¢o %he ownership

background The geographical distance between KB branches and

3O

their NCR head offices leads to an "arms'-length" _ Pelati6nship

between KB owners and branch managers, manifested in formal

management policies. On the other hand, the situation of _Bs and

to some extent, PDBbranches, is different. Tab1'e 9_ shows '_that

most RBs are managed by stockholders while few PDBs_surv_WYed

are under thesame management. Stockholders ofrural banks have

concurrent ownership of a number of other businesses, including

other_,financial institutions, manufacturing, trading and service

enterprises.

Certain _espondents claimed advantages in relationship of

the RFI with other ' businesses such as: (a),_ int_rcompany

finan_=ial "assistance, particularly with. orher financial

enterprises held by the RFI owner; (b) use of RB staff"and

facilities for transactions of_the other thrift ba_R holding; (c)

expanded client base for multiple businesses; (@) use by the

other businesses of the RFI aS depository banR and creditor and

(e) "learning experience" from the RFI enables the stOckhOlders

to .:se_ up other financial 'institutions. No corresponding

disadvantages of holding other businesses were reported by RFI

managers in the survey,

These findings on deposits, loans and ownership a_e

indicative of the more observable characteristics of RFI..,_",_,.

management. Their l_mitation lies mainly from the

Defined as cases when the president or general manager oft'he RFI is also a stockholder,

31

Table 9'

OWNERSMIP BACKGROUND OF RBs AND POB_

RBs (n=19) PDBs (n=13)

1. Ownership ana Management _Number)

a) O_ner-Nanaged 18 7b) Manager is Not a Stockholder I 10

2. Other Businesses Owned byRFI Stockholders..

a) RBs., PDB,;_Thr_ft Ba_ks, etc. 10 .3b) Real Estate/Agri-based 8 0

c) Service Enterprises 11 0d) ManUfacturing 4 0e) Merchandising 3 1

inability of .questionnaires to ca_t_re the,..subjec'cive and

judgemental elemen_ of management policymaking..

VI. RELATING _ANAGEMENTSTRUCTUREANDPOLICIESWITH OPERATINGPERFORMANCEOF RFIs"

1. Mana_emen_St_ucture and RFI Per:Eorntance

The structural features of the RFIs' organization can be

related to its performance. Several questions which can be

raised are as follows:

a) Can the larger number of staff positions in unit banks

be possibly-explained by £heir larger number of deposit

and loan customers (compSred to PDB and KBbranches)?

32

b) Is there a relationship between the size of the

organization and personnel overhead cost of an RFI?

c) Can the difference in loan approval limit among the RFI

types be possibly related to.the typical• cli,ent:-,group

serviced by the respective RFI's?

d) Can ownership and organizational structure significantly

add to an understanding of RFIfinancial status andL

performance?

The difference in number of managerial staff ,positions of

unit" banks may be possibly explained by relative diversity in

services and size of client base. An extensive deposit

•generation and lending operations as shown by the large •number of

clients served, can:justify a more extensive staffing.p}an._ This

relationship is reflected in the ratio of the number of deposit

and loan accounts to the number of staff positions in the RFI.

The result shown in Table 10 supports this claim, .PDB ,and KB

Table 10

MANAGERIAL HIERARCHY AND SCOPE-OF RFI SERVICES,Mean (Standard Deviation)

....... ====

B A N K

Rural Development KB=Branch

Number of Accounts Serviced

Per Bank Manager/Supervisor 3,995 1,985 2,450(2,336) (1,613) (1,211)

K-W One-Way ANOVA: N.S.

K-S Two-Sample Test for Each Bank Pair: All N.S.

33

branches service relatively fewer clients which may help explain

its leaner organization.

Personnel cost. is a_noverhead expense which is normally

m;"

allocated by banks • and used in priclng, ioans. H"igher •cost

structures can then influence RFIs' profi tabi Iity and

•competit_iveness in the/maf.ket,place.. Table 8.(.I.t,em I), shows that.

each RFI type has a different level of compensation ,.._pr its

staff. This may be due to inherent differences" " in

.qualifications.,r, local job.market an_i..requirem_nts."__.MeanIAgful¢

comparisons, should theh; be made only within each RFI!_category.

In. determining whether the organization size is Fe).ai_ed.,'tothe

overhead cost of an RFT., the ratio of staff posii_ions to 't_0tal

personnel i"srelated to the ratio of personnel compensation cost

to total assets in .which a positive re'_ationship _ •holds.

Although an orBanization may increase ..;_;:i,t;s`__,staff, the

corresponding resource generated by the new staff may ilower the

latter ratio.. The result, in Table 11...(item.2") shows•signi:.ficant• / ,c

posi.ti-v.erelationsh!p .- more staff .is .associated..w._i.th higher=

overhead cost - for KB branches but not for p]DBs,and ;SBs.

It seems more difficult to relate the difference in. loan

approval, limit of RFIs to the_ir lending"performance. Comparisons"

across RFI types would not be 'vaiid if there a_'e "clientele"

effects:, e.g., if traders go to RB brancheswhile farmers go "to

RBs. A Possi_;ble approach" for control ling 10an portfolio

characteristics in assessing the degree of decentr_alization of

34

Table 1 1

STAFF SIZE AND OVERHEAD COSTMean (Standard Deviation)

B A N K S

RB PDB KB Branch Significance

1. Difference Across RFITypes

Average CompensationPer Employee 28,810 35,210 59,800 049

..... (Pesos Per Year) (8,880) (24,720) (21,210)

Total Compensationas a Per Cent ofAssets, Net ofPremises 0.04 0.03 0.02 .013

(0.02) (0.03) (0.01)

2. Relationship BetweenCompensation as PerCent of Assets andStaff Positions asa Per Cent of TotalPersonnel(Kendall's Correla-tion CoefficientTau) N.S. N.S. 0.04

K-W One-Way ANOVA

lending authority is to take the ratio of the manager's lending

limit to the RFI's average loan size. The result in Table 12

shows that the degree of lending discretion given to RB and KB

_ranch" managers appear to be comparable after considering the

loan sizes normally handled by each RFI type.

35

Table 12

BRANCH MANAGER'S .DISCRETIONARY AUTHORITY ON LOAN8_Mean (Standard Deviation)

m

B A N K S

. RB KB'Branch Significance

Branch Manager's LendingLimi_t " P8,667 _P651:_667 . .O0aOl

(6,673) (837,972)

Average Loan Size P11,127 _ 93,529 .0001(10,937) .(58,005)

Manager's Lending Limitas Percent of, 135_ _1,102_-:Average Loan Size (132_) (i,491_) N,S,

.., .. ,

Insufficient data on PDB branches,

K-S_.Two-Sample Test

The theory regarding the potential conflict between

ownership and management i.s still not well-established. It

is posited that contro.l and efficiency problems ariee_if owners

delega_e, tHe"management, of the firm 'to profess:ional managers. In

such a cats,, managers, might allocaCe to themselves corporate

prerequisi=tesand still not work as.hard (as compared ¢o.an owner

who manages .the business,_himself)..In a sense,_ this: is related to

the question of the optimal number of organizational levels which

makes for corporate efficiency. By simultaneously acting as a

_anager, an owner eliminates one more layer that separates him

For example, see Horngren (19e6), Van Horne (1983);

36

from the source of revenue- the client. An often-recommended

management strategy to: cut costs and,increase_rofits +s to

reduce organizational layers separating the chief operating_/

officers from the bank's clients.

An immediate problem arises in comparing RFIs according to

the owner's role in management. As previouslY shown, most RBs

are owner-managed while most PDB and all KB branches are not.

Consequently, a comparison on owner versus management basis• , c

cannot be practically distinguished from that of a comparison of

RBs and PDBs as separate groups. For example in Table 13, it is

indicated that the return• on assets of owner-managed RFi's is

significantly lower while illsequity base is not significantly

different •from non-owner managed RFIs.

But •the same result holds .true for. RBs relati've to PDB10/

branches as well. •

...." A simultaneous evaluation of the ..owne_rsh_p., and

organizational structure variables (level) .can be made by

including them in a muli;iple regression involving return on

assets. One would expect the return on assets ratio to be

significantly related to key a balance sheet ratio like deposit

to loans ratio and productivity indicators like the ra-tio of

9/Reference to Johnson and Johnson (1985, p. 32).

10/As Table 10 (item I) shows, K-S test using RFI type as

basis yields •significance level of .0001 for return on asset.There is no significant_ difference in equity percentage for RBsand PDBs.

37

Table 1_

OWNERSHIP, STRUCTURE AND _FIPERFOPJ4ANCEMean (Standard Deviation)

B A N K

aJRB PDB Branch Significance

.. . • . .. ,,

. RFI Type Analysi_:_

a) Per Cent o_Total 95% 21_Owned-Managed

b) Return on Assets 0.02 0,08 ,0001(0.07) (0.03)

c) PerCent o{Equity to Ne_ 0.70 _ 56 .... N.S.Assets , (0.43) , (0.44)

B A N K

2. Owner Type Analysis_ Not Owner Owner-Managed ,H_nage(

a) Return on Assets 0.05 0.03 .02(0.04) (0,07)

b) _er Cent of

Equity to Net_ 0.24 0.73 N._.Ass_s ' (0.42) (0.43)

K'S Two-Sample Test

compensation to assets, The question is whether ownership and

organization levels significantly increase the explanatory power

of the regression.

The regression analysis uses return on assets ratio for 1986

as the dependent variable. A clarification on this measure is

that since Commercial bank branches transfer a large proportion

38

of branch deposit to their respective head offices, interest

income on such transferred funds'waS imputed to the branch.using

the average annual Manila Reference .Rate ('MRR). This

methodological step is equivalent to an assumption that the

branch lends out its funds to head office at the current prime

lending rate. Regression results were not significant for

average compensation, ratio of personnel compensation to assets,

deposit to loan (both net loans and the sum of net loans and"due

from head office"account) ratio., and deposits to total assets

ratio.

Organizational variables such as management by stockholder/

owner and thenumber of organizational levels showed somewhat

better results, but these were at best significant at the 10

•-percent level. Table II shows the results of this latter

evaluation wherein" the signs are as expected, i.e., more levels

in organization and ownership..by management are negatively

related to return on asset. Correlationanalysis also showed

that average.loan size and number of accounts per .bank officer

are highly correla.ted with return on assets. However, due to

the small sample in this initial data, a valid regression

analysis cannot be done.

39

Table 1_

NULTIPLE,REGRESSION OF RETURN,ONASSeTS RA_IOONFINANCIAL AND ORGANIZATIONAL VARIABLES

i" L , ,.

Regression Coefficient(t-value)(,e,ig_ificanc_..]Oyel,.of ._stati_jg).._i-

Variables in the RegressionExplar_a_oryVariable 1, 2 and 3 1 and 2 2 and 3

1. DepOst¢_to Ne_ Loans -0.003 0.0018(-0.434) (0.274)( o;,699 _ (0.786)

2, Number of Organizational -0.026 -0,027 -.0.018Levels (-1.798) (-1.831) (-1.350)

(0,087) (0.080) (0.190)

3. Dummy for Owner-Manager -0.044 -0.034(-1.495) (-1;254)(0.150) (0.2_2)

4. Zhterc_pt_ 0.1,72 0_'138 0.136(2,919), (2,350) (2.760)

: (0.008) (0.027) (0.011)_2

5_ R _ 0;153 O.07t _O.t04

6._F of Regression 2,44 .... 2;00 2.'81Significance of F-Value (0.0924) (0.1574) (0.1025)

7, Number of Cases 25 27 27

The;'¢receding table •Shows that the, organizational structure

of unit banks .are associated _ith'_ higheP_ _ financial cost

structures,., The •Ownership aspect is more ambiguous. RBS are

managed Completely by _'ts owners unlike other RFZl;types, thus

any ana)ysds,_based on the ownership variable reflects all' other

RF1-type-r_]ated,_,cha, racteristtcs. The average loan size_and

4O

number of accounts per officer variables deserve further

investigation in futu_e studies.• These variables.are potentially

indicative of economies•of scalein RFI operations which are

eventually reflected in its profit performance.

2. Management. Policies an_ REI PerfQ_m@nce

Given the preceding background on differential .man_ementI;

policies pursued by. RFIs, the analysis proceeds -along the

possible implication.s on: (a)services offered and depl_ymen_ of

staff, (b) deposit mobilization and loans performance, and (c)

overall re_urns and_"benefits tO the RFI stockholders. ....

Services OffEred aEd Staffin_

The RFI management's manpower deployment policy is a

significant indicator of the ••.relative importance .•placed, by

management'on its,service lines. A classificatign of bank staff

into•three categories, namely deposits, loans and administrative

support'showed that rural banks allocate more managers_.and staff

for loans while KB branches emphasize deposits and

administrative support (see Table 15). In this regard, POBs

operate more like KB branches rather than RBs.

• Table 15 shows the results of this analysis_ ? The

concentration..by KB and some _DB branches ondeposit servipes is

supported by the study's preceding findings on dePos_,_

management policies, Centralor area head control, of deposit

policies, regular and more frequent review of. balances, and

incentives-based deposit generation programs support the emphasis

41

Table'15

MANPOWERDEPLOYMENT POLICIES

AN_ SERVICE PERFORMANCE _ R_,IsMean (Standard Deviat{onT.

B A N K Sa

RBs PDBs KBs S i gn i f i cance

1. Number of Per_sonnel in:

....a) DeDQsits 1.17 1,31 1.93 -(i. 11) (1.14)- ('2.32)

b) Loans 2.7 0.88 0.41 -(1.61) (1 ,54) (0.84)

c) Administrative Support 3.83 4.25 4.0 -(2.57) (2 _74, ) (_.31)

2. Percentage of Personnel toTotal

a) Deposits 0.12 0.18 0.24 .02

C0,_,10) (0.12) (0.12)

b) Loans 0.31 0.08 0.05 .0001('0. 10) r ('_..,1 _,)._ r(,0.08)

• c) Administrative Support 0.42 0.56 0.56 .004

(0.!2) (0: 12)i (0.10)

3. Service Performancet

a) Deposits to Loans 0.52 _j_66 41.38 .0001(0,34_- f_, 6o.)- (se,_5:_.,

• j

b) Deposits 1;o Asse_s ,,, ....(0._23) (0.'30)

:a[ ......... ..Kruska]i-Wallis(K-W) One-Way Analysisot_variance: '

42

of KB and PDB branches on .deposits. •These management policies

may also be related to a "clientele" effect on the RFI's deposit

profiles. As shown in Table 16, thp average deposits at KB'

branches are much largerthan those of PDBs and RBs.

Table 16

MANAGEMENT MARKETING POLICIESAND DEPOSIT/LOANS PERFORMANCE

Mean (Standard Deviation)

B A N K S

RBs PDBs KBs Significance

A. Deoosit/Loans.(ThousandPesos}

1, Average Deposits 3.7 14,4 67,1 ,0001

(3.7) (10.3) (41,6)

2, Loan Approval Limit 8,7 43.8 651,6 ,0001(6.7) (41,9)

3, Average Loan Size 11.1 222,6 93,5 ,0001

(10.9) (218,6). (58,0)

a) RB vs. PDB or KB ,0001

b) PDB vs, KB N.S,r

B. C_.r re lat i on_Analxs i s

Loan Limit and AverageLoan Size: Kendall's Tau 0,24

Significance 0,16

A/K-W One-Way ANOVA

=/K-S Two-Sample Test.

43

Given the presentdata,,it is difficult to explain Why" KB

branches' lend so little compared:to the large poo:l of low~cost

deposit funds they normally generate. Several suggestions can.be

made based on Table 16. First, commercial banks encourage their

branch managers to lend to large borrowers by allowing them

higher loan approval limits Ot their level. The difference in

loan approval limits among the three RFIs is Quite significant

(per (A2) in Table 16). _, the loans clientele appears to

be of statistically different sizes for the three RFIs (see (A3).

KB and PDB •branches do "wholesale" lending while RBs take care of

"retail" loans. This suggests market segmentation and possible;. _

specialization of each RFI type in certain segments. Third, the

"large" segment of the local loans market appear to be limited

compared to the resources available to KB branches. F_9__,

smaller local loans may not be able to pay high interest if KBs

pass on the higher transaction costs associated with this

clientele.

Another interesting question is whether the average loan

size of an RFI is related to the discretionary authority allowed

to the manager. A positive relationship suggests that the

manager ,has "a role i.n"expanding, the client".ba-se of the .RF_I to

include l'arger clients.- A statistical test (Section B in Table

16) correlating the RFI manager's loan approval limit to the

RFI's actual average loans per account, was done • f̀or RBB (being

the only case with sufficient data). The" result indicates a

slight tend_cy of positive relationship between the manager's.

lending limit and the average loan size of the bank. For' the

44

limited sample of this study, RBs which allowed higher lending

authority to their managers are able to reach out to larger

borrowers.

An appropriate summary is a response on the comment in

Lamberte (1987) that disparity in deposits performance by KB

branches may be due to wide discretionary authority enjoyed by KB

branch managers for raising deposits but not for originating

loans. The results in this study suggest the opposite. Deposit

policies and initiatives are made at head office and the branch

manager is given a lending limit which is far higher than the

local loans market. This implies that head office policy is the

main factor which determines branch operations. Once the head

office mandates a deposit priority, the performance of the branch

is primarily determined on this basis. The higher lending limit

given to managers may even serve as a deterrent because it

appears to be set in relation to head office lending norms (e.g.,

designed to avoid high transaction costs) rather than 'the market

faced by the branch office.

Deoos_Mobilization and Loans Performance

The limited survey data on savings campaigns can be used to

evaluate whether such efforts are potentially fruitful. SinCe

KBs appear to have far greater deposit generation capacity, the

test of any relationship between savings compaigns and deposit:

generation will be done only for. RBs and PDBs. Data is available

on savings campaigns for these RFIs, shown in Table 17.

45

Table 17

SAVINGS CAMPAIGN AND DEPOSITPERFORMANCE FOR RBs AND PDBs

B A N K S

RBs PDBs Significance

1. Savings Campaign (Per Centof RFIs) N.S.

a) With active campaign 71.4 68.8

b) No active campaign 28.6 31.2

B A N K S

.....With Campaign No Campaign

2. Deposits to Assets.(Mean) 0.69 0.37 .02(Standard Deviation) (0.31) 0.29

i

Chi Square

From this table, the propensity to undertake deposit

campaigns is statistically equivalent for RBs and PDBs. The

resource generation performance of RFIs which undertake deposit

campaigns is superior tothose which do not, for the survey

sample. Meyer (1987, p.6) conjectured that some RBs may have

problems in theiraggressivedeposit mobilization program due to

image problems among customers. The evidence in this study

indicates otherwise. RFIs undertake such campaigns through a

combination of person-to-person promotion, staff incentives and

prizes and give-aways to customers. Evidently, deposit campaigns

46

in the rural sector are not impersonal in approach and well-

managed RFIs can think of an appropriate approach to avoid the

problem cited by Meyer (1987)..

The wide disparity in deposits generation across RFI types

carries over to loans performance. Rural banks lend more

relative to deposits while PDB and. KB branches generate far more

deposits than they can (or are prepared to) lend in. the local11/

area. Excluding the extreme case of KB branches once more,

the relative deployment of bank staff for loans and deposit

functions is a significant indicator-of the financial (loans to

deposit) performanceof the RFI. This can be seen in Table 15 -

the .loans to deposit ratio is significantly different for pairs

of RFI types. Hence while the loan approval limits of KB

branches, and PDBs far exceed those of RBs', there are just not

too many sizeable loans in the rural areas. Given this loan

market, the relative number of staff assigned to these functions

is actually a good indicator of the financial portfolio of RFIs.

Ownership.. Management and ODeraJ_tnq Performance

The operating strategies_followed by the management of RFIs

can be expected to influence their summary operating performance.

For example, would the lackof, deposits to finance lending wby

RFIB lead to lower rates of .profitability? Is the concentration

11/The result for PDBs may be somewhat contaminated since

some PDBs in the survey are unit banks while others arebranches.

,12/Significance of K-S Two-Sample Test are for RBB and PDBs:

.002: for RBs and KBB: .0001.

47

of staff on deposits function among PDBs and KBs associated •with

a lower overhead cost, structure and higher profitability? These

questions may be addressed using the profitabi'lity ratios as

starting point,

A technical questionneeds to be resolved in the case of KB

branches. When deposits for these banks far exceed their loans,

the branch income statement would show large deposi{ iriS.crest

expense but. minimal lending interest income. The_e losses are

aggregated in branch books leading to a negative equity position

for some branches. From•the viewpoint of the head office •, such

problems are entirely immaterial to operating ,decisions and

disappear • with periodic consolidation of branch accounts for

bank-wide financial reporting. For some commero_del banks,

branches are evaluatedbased on "contribution margins", _in_" which

a traMsfer pricing policy is implemented. The _cheme _nvolvesL. i

imputing interest income on transfers by branches of their

deposit funds to head office. In this study, return on assets is

calculated for KB branches (due to the absence of "equity")and

branch "profits" include an imputed income based on the Manila

Reference Rate (MRR). In KB branches which use transfer prices,

the profit figure is not adjusted. A comparison of return on

assets is shown in Table 18.

' 48

Table 18

PROFITABILITY PERFORMANCE OF RFIsPercent (Standard Deviation)

B A N K S

RBs PDBs KB Branches Significance

1, I_Leturn on Assets, Netof PcemiEes (Percent)

a) All three RFI types 1.96 7,66 2.41 .0001(7.02) (3.07) (1.78) "

b) Pairwise Comparisons

i) RBs and PDBs - - - .0001li) RBS and KBB - - - N.S.

iii) PDBs and KBs - - - .002

2. Adjusted Return o_AssetB,Net of Premises (Percent)(Net Income +ManagementFees + 2.67 7,70 N,A. .00iDirectors Fees) (7.45) (3.04)to Assets, Net ofPremises

K-W One-Way ANOVA

K-S Two-Sample Test. N,S, at 0.20

With the previously described procedure for adjustment of KB

branch income, the highest return on asset (ROA) ratio can be

found in PDBs followed by KB branches and R Bs. It is likely

that the ROA for KB branches is understated to the extent that

their actual income includes a margin over MRR and the income

imputation method is applied on •ending branch ba]ances for

transfer to head office without considering turnover.

49

It can: be =recalled.that,these rural-banks .....and private

developmem;b banks differ on the basis of the. owner.ship,, variable,

Since inmost oases, RB stockholders also manage the. bank, thei,r

personal (or;.family).income consist@ oR both net profits.of the

bank _nd management/directors' fees. _ If return on .assets is

adjusted ..,to .include thi_type ' of .bank, expert, as for bqth RFI

types, the--adjusted returns are still statistica]l_ different'

(.see Item. 2), i..e., rural banks have lower returns, jZt shouldbe

pointed, out thatthe key performance variable is the returns on

equity_- particularly in evaluating the incentive, and success of

owners, _in,,their profit seeking under_aking. Unfortunately, the

lack of data did notat.low the application of this measure in

this study,

The ,analysis of the incentives to owners ana interaction

with the management choices is not complete without consideration

of DOSRI loans. • From the responses given by unit banks regarding

the "advantages" of having stockholders with other business

interests, it shows that owners can obtain further incentives or

benefits_ through (DOSRI) loans to their other businesses. In

concept, the _rof_t to_ the.owner-managers, of RFIs is. equal to the

sum of the net.'income of the-RFI plus their management/directors

fees • plus advantages gained through any below-market interest

rates ow,DOSRI, l_ans_,.It i-s in•this oontext that Tolentino (1987)

points to the possibility that rural bankers capture,the gains

from subsidies t_;.(_ther intended beneficiaries ("The rural bank

failed, but the rural banker got rich").

What •analysis can show the effects of DOSRI? One approach

may be to obtain a correlation of return on assets to the amount

of 'DOSRI loans by RFIs. If DOSRI loans are concessionary ih

favor of the other businesses there should be a negative

correlation between these -two variables. The correlation

expected is reversed if the loan is concessionary in favor of.the13/

RFI. The result: Kendoll's COrrelation Coefficient Tau of

+ 0.06 has a significance level of only 0.38. This result could

be due to the intervening effect of RFI"s expenses in deriving

return on assets . Hence another approach may be to look at the

average gross interest (income) yield on loans and oorrelate this

with the size of DOSRI for each RFI. Unfortunately, the data in

the survey is not sufficient to allow an analysis of the yields

on loans of the RFIs. Also, as previously explained, the results

may differ when • returns on equity is used. •These Would be

interesting subjects for future inquiry.

VII. CONCLUDING REMARKS

The results presented in the.paper, while preliminary, show

the relevance of analyzing.organization and management structure

toward understanding the differential characteristics !_a_d

performance of the three types of RFIs. Unit banks - RBs and:_-to

some• extent,' PDBB - are organized for a balanced 'o@Terimg :of

_eposit and loan services within their own. regul,atory

re'Strictions. .These RFIs tend to show more organiz.ati_nal

13/The range of Tau is -1 to +1.

51

levels, ,,hierarchical in staff relationehips, and organizational

positions compared to branches_of KBs, The authority ,given ,to

RFI managers appear to be more of a function of bank size (and

so, RFI type) than management policy, Many decisions remain

centralized, especially on key deposits and lending policy

aspects, Management structure was also found to be associated

with overall RFI performance. The hierarchical and multi-level

structure of unit banks is related to the higher overhead cost

structure and lower return on asset performance of these banks.

,,t Many _of the conclusions derived i n._,this sl_udy:,.support

previous findings about =the operations of RFIs and how one_RFI

type differs from another, The major ones are now cil_d by way

of summary, Eirst, like rural banks and private ;development ,

banks, unit banks operate more like community banks, serving the

deposits _ tending,needs of the local area.' _Commer¢,$a] bank

and deyelopment bank, branches are not;as oriented to community

banking, being Subject to central decisionmaking authority on

deposit and loans policy. At this time, the evidence appears to( r

indicate the dominance of two factors: the profit motive of

RFIs and the effects of a regulatory/incentive structure

designed to favor loans to urban, large-scale ventures as cited

in Tolentino (1987). Hence, KB and PDB branches are

predominantly operated as deposit-taking branches, Second. it

was found out that RFIs can conduct savings mobilization

campaigns and substantially benefit from • them. The range of

operating strategies available to management in this regard

52

appears sufficient to overcome common concerns'associated with

small.,,banks. ..

Third, the issue of "small" lending portfolio by commercial

banks in the local area may be associated with a combination of

factors like: (a) a "clientele" effect; (b) poorly developed

loans market; and (c) a continuing signal from bank head

offices that large loans are encouraged. The higher salary and

cost structures of commercial banks appear to be a deterrent for

developing their branches into community banking. These factors

also make larger loans a prerequisite for profitability ,of loan

accounts. Eourth, given the current state of small loans market,

management's deployment of its staff into loans and deposit

functions already serve to indicate whether it emphasizes loans

or deposits. Policymakers interested in periodic assessment _,of

the banking functions of RFIs can avoid the high costs of

financial audits by looking instead at this surrogate indicator.

Fifth, ,the ownership aspect of RFIs appears to be a key

variable in management decisions and RFI performance.

Policymakers should view its proposed regulatory schemes in

relation to the incentives to owners in a total context, in this; 6"

respect, innovations need to be made regarding traditionalt

ratios. This ownership variable is the counterpart of the head

office in the case of commercial bank branches. Policy analysts

should interpret regulations in the light of the wider options[

available to commercial banks toward a better understanding of

intended effects.

53

Recommendations are not,,,as easy.to specify.- ..... )What ,clearly

emerges is',an impression'_that a number,.of current Conditions must

be addressed • in order to bring about an_,organized deve]ol_ment of

the .ru,ra_, financial s:ys,tem.. -One,_ condit,ionis._¢he competition of

_'_restriGted" rural banks_ and "centralized" commercial bank

brancbes, Profit_seeking-_bshavior (and.restrict,ions or lack o.f

it) made,,, ruPalbanks more,,.community-based and commercial bank

branohes more centralized•in loans..

An initiative to make thrift banks establish corresponoen¢, _ :. , :,-.' "_..', , .. ,:, "

relationships with commercial banks may enable them to operate• .'",,.

more like branches. On the other hand_ the exhortations by

pol,icymakers for comm_,cial bank branches to,be more "community

and-small., bu_s-orA_ted", will not necessar,ity.be foil,owed, by

RFIs unless , rural loans grow larger (e.g.;, local businesses

develops'):. Alt_rnati've_y,,_ commercial ban_ ca n_,,ac_ud_re more

_h,rift banks_,in-or_er<to,Bnjoy these banks' lower, cost:_structures

a,_d, be more "community'oriented ." The,,#,indings here can-also be,.

appl-ied te branch bank.inQ,....In.a recen__news _-tem, _ .,,,_a banker

suggested "that central _ank-s.houqd g.ive incentives for, banks, to

open, • brar_hes ,in the, countryside. _Thi.s ,study suggests that

Central -,Bank should..fi,rst reexamine.the role of branch •baRks in

thec.ountr,=yside in the.o_text_of , the r_n.ge of services prior to

..granting _"im_ent:i "' yes ,.:.:"

14/"PCIB seeks easier bank branch rules, "The,,Joupr_al.

August 30, 1988.

54

Policymakers also need to look into innovative means of

monitoring RFI performance along key variables that reflect ,the

development of rural finance. A, better understanding of branch

operations and the actual iincentives tO owners and. ,managers of

rural financial institutions should enable, .,government and

analysts to understand the impact of planned regulations, and

credit programs. This is one area wherein technical,, development

is needed by researchers and analysts to enable them ,.to

understand, issues such as whether financial subsidies intended

for certain, beneficiaries are "captured" by the banker, in the

rural areas.

There are also significant,, limitations of the study,

,foremost of .which is the fact that t_e. management aspects

analyzed, in the study are mainly of the "quantit_ative"_ variety.

For example_ the measures of structure, levels and manager-staff

distributions variables do ,not necessarily describe ,fully the

various dimensions of organizational structure. The methodology

• adOpted here merely addresses the need to combine the results

across all RFIs surveyed into a "sectoral" pictur_.;_ This me=n,s

that there are research possibilities involving case studies

of these RFIs. Such microstudies can address the other

interesting but more qualitative management and ,,organizational

features. The framework offered in this. study is useful in _Gase

studies as starting point and basis for comparison of results

with a "sectoral" management study.

55

Another llmltatlon is the apparent lack of reference to

related issues that malnly affect management structure For

example results of studies o# borrower and lender transaction

costs, special government credit programs transfer prlclng

pollcles and savlngs capaclty and moblllzat_on in rural areas may

lndlcate influences In the management of RFIs There are

contemporaneous studles on these aspects which need to be

evaluated

REFERENCES

Horngren, C Cost_ccountlnq A Mana_erlal EmOhasls 5thEdition Prentice Hall 1986 ,

Johnson and Johnson Commenclal Bank._Manaaeme_ _ Chicago,lllinols The Dryden Press 1985

Koontz, _'Donnel and Welhrlch Managemen_ 8th Edition NewYork MoGraw H_ll, 1984

Lamberte, Marlo B Comparative Bank Study A BackgroundPaper Working Paper Series No 87-04, Philippine institutefor Development Studies April 1987

Meyer, Richard L Rural Finance in the Philippines RecentChange_ and Priority Issues, Economics and SociologyOccasional Paper No 1358 The Ohio State Un_verslty June1987

Porter Michael CQm_e_j._lve Strategy Techr_aues for Analvzln_

_duStr_es and Co_JnDetl_o.rs Free Press, 1980

Sorla_O, Emienuel V Bus_ness Policy In_n As_an Context Textand_Cases 2nd Edition Slnagtala Publishers, inc 1976

Tolentlno, V Bruce J Current Imperatives andDevelopments _ in Phll_pplne Agricultural Ppllcy, Economicsand Sociology Occasional Paper No 1324, The Ohio StateUniversity, March 1987

Van Home, J Financial Management and_l_cv. 6th EditionPrentice Hall, 1983

56

PIDS WORKING PAPERS

W.P. No. 88-01 A General Assessment of W.P. N0 88-10 Financing the Budget,'Foreign Trade Barriers to Deficit in, a Small Open

Philippine Exports. Economy: The Case of theErlinda M. Medalla Ph_ppines, 1981-86.(P23.00) Ma. Socorro S. Gochoco

W_P,No. 88-02 Economies of Upland Re- (P'26.00)source Depletion: Shifting W.P. No. 88-11 The On-site and Down-Cultivation in the Philip- stream Costs of Soil Ero-pines. Marian S. delos sion. Wllfrido 13. Cruz/Angeles (tw23.00) Hermin_a A. Francisco/

W_. No. 88-03 The Size, Financing and Zenaida Tapawan.ConwayImp act of the Public (P61.00)Sector Deficit, 1975-1984. W.P. No. 88-12 A Review of Policies Ira-Rosario G. Manasan pinging on the Informal(P17.00) Credit Markets. Meliza H.

W.P. No. 88-04 An Analysis of the Role of Agabin (P30.00)Pawnshops in the Flnan- W.P. No. 88-13 Flexible Functional Formcial System. Mario B. Estimates of PhilippineLamberte (-P'14.00) Demand Elasticities for

W.P. Nol 88-05 The Financial Markets in Nutrition .Policy Simula-Low-Income Urban Corn- tion. Agnes Quisumbingmunities: The Case of (?50.00)

Sapang : Palay, Mario B. W.P. Nol 88-14 Political Economy ofLamberte [IP'26.00) Credit Availability and

W.P. No. 88-06 Informal Savings and Financial Liberalization:Credit Institutions in the Notes on the Philippine"Urban Areas: The Case of Experience. V. Bruce J.Cooperative Credit Unions. Tolentlno (PI$.00)

Mario B. Lamberte and W.P. No. 88-15 Rural Deposit MobtIiZa-JovenZ. Balbosa tion in the Philippines,(P40.00) 1977-1986. Rhenee Blanco

W.P. No. 88-07 The Manufacturing Sector and Richard Meyerand the Informal Credit ¢1_17.00)Markets: The Case ofTrade Credits in the Foot- W_P. No. 88-16 Comparative. A_eementwear Industry. Mario R Structure and InstitutionalPerformance in Rural

Lamberte and Anita Abad Banking Institution.Jose (P35.00) Cesar G. Saldafia (_0.00)

W.P. No. 88.08 Japan's Aid to ASEAN:Present Realities and W_v. No. 88:17 Borrow Transaction costsFuture C'hallenges. and Credit Ratldning inFilologo Pante, Jr. Rural Financial Markets:0PI5.00) The Philippine, ,, q2ase. '

Virginia G. Abiad, CarlosW.P. No. 88-09 The Effect on an Exchange E. Cuevas and Douglas H.

Rate Devaluation on a GrahamSmall Open Economy with (P27.00)an External Debt Over-

hang.JosefT. Yap W.P. No. 88-18 Transactions Costs. and• the Viability of Rural

(P9,00) Financial Intermediaries.Teodoro S. Untalan andCarlos If. Cuevas (P24.00)

57

W.P. No. 88-19 Credit Rationing Under a W.P.No. 88-25 The Urban InformalCreditDeregulated Financial Markets: An IntegrativeSystem. Ma. Lucila A. Report. Mario _ LamberteLal_r and Douglas H. fP19.00)

Graham CP13.00) W.P.No. 88-26 The 1989 Program ofW.P.No. 88-20 The Analysis of Savings Government Expenditures

Behaviour: The Case of in Perspective. Ro_rioRural Households in the G. Manasan (PIT.00)

Philippines. Jocelyn Alma W.P. No. 88-27 A Review of InvestmentRodriguez and Richard L. Incentives in ASEANMeyer (1"23.00) Countries.. RomFio G.

W.P. No. 88-21 The Demand for Funds Manasan (YIT.00)Among Agricultural House-holds in the Philippines. W.P.No. 88-28 Science and Technologyand Economic Develop-Raquel Clar de Jesus and ment. Mario B. LamberteCarlos E. Cuevas (P28.00) (P13.00) ,

W.P. No. 88-22 Funds Transfer Operation: W.P.No. 88-29 The Impact of Trade, TradeBoon or Bane to the policy and External ShocksViability of Rural Finan- on the Philippine Economyclal Intermediaries. Julius Based on the PIDS-NEDAP. Relampagos and Mario Macroeconomitric Model.B. Lamberte fP17.00) Winnie M. Constantino and

Josef T. Yap(el3.00)

Copies may be obtained s .....

RESEARCH INFORMATION STAFF (RIS)PHILIPPINE INSTITUTE FOR DEVELOPMENT STUDIES

ROOM 307, NEDA SA MAKATI BUILDING106 AMORSOLO STREET, LEGASPI VILLAGE

MAKATI 1200, METRO MANILA, PHILIPPINESTELS: 86-57-05 / 88-40-59